text:Template Journal 7/17/09 1:25 PM Page 433
Optimization of shovel-truck system for
surface mining
by S.G. Ercelebi*, and A. Bascetin†
Synopsis
In surface mining operations, truck haulage is the largest item in
the operating costs, constituting 50 to 60% of the total. In order to
reduce this cost, it is necessary to allocate and dispatch the trucks
efficiently. This paper describes shovel and truck operation models
and optimization approaches for the allocation and dispatching of
trucks under various operating conditions. Closed queuing network
theory is employed for the allocation of trucks and linear
programming for the purpose of truck dispatching to shovels. A case
study was applied for the Orhaneli Open Pit Coal Mine in Turkey.
This approach would provide the capability of estimating system
performance measures (mine throughput, mean number of trucks,
mean waiting time, etc.) for planning purposes when the truck fleet
is composed of identical trucks. A computational study is presented
to show how choosing the optimum number of trucks and optimum
dispatching policy affect the cost of moving material in a truck-
shovel system.
Keywords: Open pit mine, equipment selection, dispatching,
linear programming, closed queuing network theory
Introduction
In a surface mining operation, a materials
handling system is composed of loading,
hauling and dumping subsystems. Effective
and efficient materials handling systems can
be developed only through a detailed consid-
eration of these subsystems in a systems
analysis framework. The transport of material
from production faces to dumping sites is
accomplished by rail, truck, belt conveyor or
hydraulic transport. Shovel-truck systems are
most common in open pit mining. Two
available techniques to analyse these systems,
linear programming and queuing models, are
used and compared in this study. The most
important factor in every operation is
profitability. Productivity of equipment used is
an important factor of profitability. Profitability
can be increased by optimization of the
equipment combination used. Therefore the
first goal in these models is to maximize
productivity and hence increase production,
which in turn will result in cost reduction.
Studies conducted for the truck allocation
were carried out by several authors. Muduli
and Yegulalp (1999) studied the modelling
The Journal of The Southern African Institute of Mining and Metallurgy
T
e
c
h
n
i
c
a
l
truck-shovel systems as a closed queuing N
network with multiple job classes. Soumis et o
al. (1989) discussed the evaluation of the new
truck dispatching in the Mount Wright mine t
using linear programming. Sgurev et al. e
(2003) studied an automated system for real-
time control of the industrial truck haulage in
open-pit mines. Alarie and Gamache (2002)
studied the overview of solution strategies
used in truck dispatching systems for open pit
mines. Nenonen et al. (1981) used the
interactive computer model for truck/shovel
operations in an open pit mine; Ramani (1990)
studied the haulage system simulation
analysis in surface mining. Barnes et. al.
(1972) studied the probability techniques for
analysing open pit production systems.
Carmichael (1986) applied cyclic queuing
theory to determine the production of open-cut
mining operations, and Koenigsberg (1982)
used in his study some concepts of queuing
theory. Shangyao et al. (2008) developed an
integrated model that combines ready mixed
concrete (RMC) production scheduling and
truck dispatching in the same framework.
Sabah et al. (2003) present a methodology
based on the queuing theory, which is
incorporated in a computer module to account
for the uncertainties that are normally
associated with the equipment selection
process.
Proposed models
Optimum number of truck assignments
to shovels (by employing closed queuing
network theory)
In a shovel-truck model, trucks cycle between
their assigned shovels and dumps or crushers,
* Istanbul Technical University, Mining Engineering
Department, Maslak, Istanbul, Turkey.
† Istanbul University, Engineering Faculty, Mining
Engineering Department, Avcılar, Istanbul, Turkey.
© The Southern African Institute of Mining and
Metallurgy, 2009. SA ISSN 0038–223X/3.00 +
0.00. Paper received Oct. 2008; revised paper
received May 2009.
▲
VOLUME 109 JULY 2009 433
text:Template Journal 7/17/09 1:25 PM Page 434

Optimization of shovel-truck system for surface mining

over haul roads. When calculating cycle time for a truck, the on. P(N,0,…,0) may be obtained from the requirement that
time taken to spot and load, haul, dump and return needs to the sum of the probabilities equals 1, such as:
be considered. The nature of these activities includes
[4]
variability in the cycle time. Trucks do not normally arrive at
the shovel to be ‘serviced’ in a predictable manner, nor does
it take exactly the same time for the shovel to service each [5]
truck. The interaction between the randomness of inter-
arrival times of trucks and the shovel service time results in
either trucks to queling at the shovel or the shovel being idle For N cycling units,
while waiting for a truck to arrive (Elbrond, 1990).
Ore or waste is moved from shovel locations along a [6]
network of haulage roads, to several dumping or crusher
stations. Through extensive time studies in the field, data are
N = number of trucks
collected on the load times, the truck travel times, waiting
M = number of phases
times for the trucks at the shovel and at the dump location,
μi = service rate at i th phase
and the truck dump times. Statistical distributions are fitted
The probability that a phase is working (phase
to the observed data. These distributions permit the random
utilization) is:
selection of event times for the defined sequence of
operations. [7]
The queuing theory calculation is fast and simple. In
truck dispatching this could be advantageous because
The expected number of trucks in the queue at the i’th
forward estimates of waiting times are important information
phase is:
for the dispatcher. However, most mining applications are
highly complex and accurate modelling results in complex
[8]
queuing models that have no direct analytic solution. Usually,
cyclic queuing models are solved by assuming that arrival
and service mechanisms are Markovian. Approximation of
The expected time that a truck spends in the queue at the
times of loading hauling and dumping, with exponential
i’th phase is:
distribution is a typical example of this situation.
[9]
A typical cyclic queue in an open pit operation may be
considered to consist of four phases (Figure1): Θ = η i μ i ; number of trucks being serviced at the i’th
1. The shovel (service; loading the trucks) phase during one unit of time.
2. The loaded haulage road (service; travelling loaded) The expected time that a truck spends in the i’th phase is:
3. The dump site (service; emptying the trucks)
[10]
4. The empty haulage road (service; travelling empty).
Since traveling, loading, waiting and dumping times are Then average total cycle time for a truck to complete M
exponentially distributed, service rates are the inverse of phases becomes:
mean service times. The cycle times of the trucks are
calculated as: [11]
The average cycle time = load time + dump time +
queuing time at the shovel + queuing time at the dump +
loaded haul time + empty haul time.
In the cyclic model the number of possible states for N
cycling units (trucks) and M service centres:

[1]

When phase 2 and 4 are transient phases such as

travelling phases, the steady state probabilities are solved in
terms of one of the unknowns P(N,0,…,0), (Carmichael
1987):

[2]

[3]

(n1, n2, K, nM) shows the possible states, which means

that there are n1 units in phase 1, n2 units in phase 2 and so Figure 1—Phases of shovel-truck system
▲

434 JULY 2009 VOLUME 109 The Journal of The Southern African Institute of Mining and Metallurgy
text:Template Journal 7/17/09 1:25 PM Page 435
Optimization of shovel-truck system for surface mining
Production over a given time period of interest (typically
one shift) can be calculated by the number of loads that
trucks take to the dump:
[12]
where N is the number of trucks in the system. Also
production may be calculated from:
[13]
ηshovel is shovel utilization and μshovel is shovel loading
rate.
For shovel-truck type operations, the minimum unit cost
of moved material is the main concern. When the cost is of
prime importance, a trade-off is sought between the cost of
idle time of the shovel and the cost of providing extra trucks.
The solution yields the optimum number of trucks of any
given capacity that can be assigned to a shovel.
For an operation involving single shovel and N trucks,
the total hourly cost is C1 + C2 N, where C1 is the cost per unit
time of shovel and C2 is the cost per unit time of a truck. Both
costs include ownership and operating costs. So the total cost
for unit production can be found from:
[14]
Once the unit production cost is found for a different
number of trucks, the cost can be plotted vs. the number of
trucks, and the optimum truck number, which minimizes the
cost, can easily be determined.
Dispatching of trucks to shovels (by linear
programming)
The linear programming model assumes no truck queuing
under ideal conditions and guarantees maximum shovel
utilization. LP model minimizes the number of trucks
required for shovel coverage without truck queuing and is
equivalent to maximizing overall production rate. The LP
function to be minimized is the total number of trucks
required to maintain all rate-limiting nodes at their maximum
production rate, subject to continuity, rate limiting, and non-
negativity constraints. A pit is viewed as a fixed number of
sources (load points) and sinks (dump points), called nodes,
connected by valid transaction routes called paths. Shovels
dump sites, and crushers are the nodes in an LP model.
Roads are the paths between nodes. Some nodes are
considered rate limiting (shovels), whereas others (waste
dumps) are assumed capable of handling all transactions.
If there are N nodes in a pit, then there are N*(N-1)
directional paths interconnecting these nodes, although some
paths may not used under normal operating conditions. For
example, dump-to-dump and shovel-to-shovel are never
used. Also some shovel-to-dump paths may not be feasible
because of topography or non-existing roads, and not used.
The general problem of allocating resources (trucks) to
activities (node transactions) can be formulated as follows:
(White et al. 1982)
[15]
The Journal of The Southern African Institute of Mining and Metallurgy
T
The objective function minimizes the number of trucks on e
the road + number of trucks at shovels (source points) +
number of trucks at dump sites (sink points).
c
Subject to the constraints of continuity: h
[16] n
This means balancing equations at each node such as:
i
incomings-outgoings = 0 c
and limiting rates at sources: a
[16] l
Meaning; Σoutgoings=1/loading time
and, finally, non-negativity constraints: N
PI ≥ 0 o
where:
NT = performance functional (number of trucks)
t
NP = number of feasible paths e
NS = number of non-rate-limiting sinks
NO = number of rate-limiting nodes
PI = average rate over path i (trucks/min)
TI = average travel time over path i (min)
PJ = sum of all sink input rates (trucks/min)
SJ = average sink processing time (min)
RI = limiting node rate (trucks/min)
The LP solution yields the desired path capacities in
trucks/ unit time for each valid path.
Case study
Mine information
In this case study, some research has been carried out to
optimize the material handling system for overburden
removal of an open-pit coal mine. The coal mine is situated
about 65 km north of Bursa, in western Turkey, and has
been in continuous operation since 1979. Currently, the mine
supplies coal to Orhaneli power plant unit (1 × 210 MW) and
to domestic users. In this case, the overall measurements of
the mine should be designed again in terms of transporting
system, equipment fleet, etc. Some technical parameters of
the working site, which affect the system, have been
researched thoroughly and summarized below in detail
(Bascetin 2002; Bascetin 2004):
The present extent of the open pit is 5 500 m by 3750 m
and a total of 75 m of overburden removed in three 15 m
high mine benches. The face inclination on individual
benches is 75 degrees, while overall pit slope is 45 degrees.
The mine will be worked over 18 years at the rate of one shift
(12 h) per day, seven days a week for 300 days per year, and
the scheduled operating time is 3600 h/year. The equipment
in the inventory reported are given, briefly, in Table I.
Optimization study
The overburden removal subsystem is analysed for the
purpose of minimizing the truck fleet size and the minimizing
unit cost composed of loading and hauling. The overburden
removal subsystem employs two 15 yd3 and two 10 yd3
shovels both with 77 ton trucks. The mine has two dumping
sites. The shovel truck system requires about 9 million m3
overburden removal yearly. The remaining 6 million m3 is
handled by dragline. The present operation of the shovel
truck system, is a closed system as shown in Figure 2.
▲
VOLUME 109 JULY 2009 435
text:Template Journal 7/17/09 1:25 PM Page 436

Optimization of shovel-truck system for surface mining

An example of queuing calculations, from shovel S22 to
Table I
dump site H6, is given below for 4 trucks allocated to the
Equipment on the mine shovel. There area total of 35 states, and corresponding state
probabilities are given in Table IV.
Equipment Number Explanation
Using Table IV, system performance measures can be
Drilling units 3 9’’ DM50 calculated.
Rope shovels 2 Marion 191 MII -11,4 m3 (15 yd3)buckets Utilization of the shovel,
(2 in waste) η1 = 1–∑P(0, n2, n3, n4)
Front shovels 5 PH 1900 AL-7.64 m3 (10 yd3) buckets = 1–∑(state 1, 2, 3, 5, 6, 8, 11, 12, 14, 17,21, 22, 24,
(4—currently 2 in waste and 1 in ore)
27, 31 possibilities)
Dragline 1 1260-W Bucyrus-Erie-25 m3 bucket
= 1–0.384
Trucks 50 Caterpillar 777-77 tonnes—(27)
Komatsu 785-2, 77 tonnes—(13) = 0.616
Komatsu 785-2, 50 tonnes—(4)
Komatsu HD 465-3—(6, coal trucks)
Bulldozer 9 Komatsu D355A-410 hp—(5) Table II
Caterpillar 81—(3)
Cat 824 wheeled dozer—(1) Lengths and travelling times for possible paths
Loader 7 Caterpillar front-end loader—(4)
Volvo front-end loader with 5.5–6 m3 buckets—(1) Path Path length (m) Travel loaded (min) Travel empty (min)
Champion-120 hp-(2)
S11-W5 780 2.5 1.5
Grader 1 Caterpillar—(275 hp)
S11-W6 1205 5.4 3.9
S12-W5 2615 6.5 4.6
S12-W6 1068 4.7 3.0
S21-W5 1500 6.0 4.6
S21-W6 1874 8.0 5.3
S22-W5 1337 5.7 4.2
S22-W6 1753 7.5 5.0

Figure 2—Shovel-truck system as closed system

In order to optimize the shovel truck system, two aspects

are considered in order.
➤ Optimum number of truck assignments to shovels (by
employing closed queuing network theory)
➤ Dispatching of trucks to shovels (by linear Figure 3—All possible truck paths
programming).
Optimum number of truck assignments to shovels
Table III
All possible paths are analysed by the closed queuing
network model, which is explained earlier. Path lengths and Cost of shovels and trucks
travelling times are shown at Table II. For this purpose all
possible truck paths to shovels are shown in Figure 3. Equipment Ownership Operating Total
cost $/h cost $/h cost $/h
Manoeuvering + loading times of the 77 tons trucks are
2.03 and 3.0 minutes for 15 yd3 and 10 yd3 shovels respec- 15 yd3 shovel 25.33 100.00 125.33
tively. Truck emptying time at waste site is 1.5 minutes. The 10 yd3 shovel 22.67 85.00 107.67
77 ton truck 16.67 40.00 56.67
cost of the shovels and trucks is given in Table III.
▲

436 JULY 2009 VOLUME 109 The Journal of The Southern African Institute of Mining and Metallurgy
text:Template Journal 7/17/09 1:25 PM Page 437

Optimization of shovel-truck system for surface mining

T
Table IV e
System states and corresponding probabilities
c
= 19.478 minutes h
State no. System state Coefficient Prob. (state)
Production = 17.453 tons/minute n
1 0 0 0 4 .321502 .005784 i
2 0 0 1 3 .385803 .006941
c
3 0 1 0 3 1.929013 .034704
4 1 0 0 3 .771605 .013882
a
5 0 0 2 2 .347222 .006247 l
6 0 1 1 2 1.736111 .031234 The above calculations are carried out for 2,3,…,6 trucks
7 1 0 1 2 .694444 .012494 and results obtained are summarized in Table V, and cost per
ton vs. number of trucks is plotted in Figure 4. N
8 0 2 0 2 4.340279 .078085
9 1 1 0 2 3.472223 .062468 The results of the queuing network solution to determine o
10 2 0 0 2 1.388889 .024987 the optimum truck number, which minimizes the unit cost t
hauled for all possible paths along with shovel utilization and
11 0 0 3 1 .208333 .003748
production are found in Table VI.
e
12 0 1 2 1 1.041667 .018740
13 1 0 2 1 .416667 .007496 As seen from Table VI, from S11 (shovel 11) to W5
14 0 2 1 1 2.604167 .046851 (waste 5) with 3 trucks, from S12 to W6 with 5 trucks, from
15 1 1 1 1 2.083333 .037481 S21 to W5 with 6 trucks and from S22 to W5 with 4 trucks
16 2 0 1 1 .833333 .014992 result in the lowest cost employing 18 trucks in total.
17 0 3 0 1 4.340279 .078085
18 1 2 0 1 5.208334 .093702 Dispatching of trucks to shovels
19 2 1 0 1 4.166667 .074961 Figure 3 shows all possible feasible paths for Orhaneli open-
20 3 0 0 1 1.666667 .029985 pit mine for overburden removal. In Figure 3, trucks are free
21 0 0 4 0 .062500 .001124
to travel between shovels and waste sites. They are not
22 0 1 3 0 .312500 .005622
assigned to a single shovel. In this way, after a truck dumps
23 1 0 3 0 .125000 .002249
its load, it may travel to any shovel for the next load. LP
24 0 2 2 0 .781250 .014055
25 1 1 2 0 .625000 .011244
26 2 0 2 0 .250000 .004498
27 0 3 1 0 1.302084 .023425
Table V
28 1 2 1 0 1.562500 .028111
29 2 1 1 0 1.250000 .022488 Summary of system measures from shovel S22 to
30 3 0 1 0 .500000 .008995 dump site H6
31 0 4 0 0 1.627605 .029282
32 1 3 0 0 2.604167 .046851 Number Waiting time (min) Shovel Production Unit cost
33 2 2 0 0 3.125000 .056221 of trucks utilization (tons/min) (¢/ton)
34 3 1 0 0 2.500000 .044977 shovel dump
35 4 0 0 0 1.000000 .017991
2 0.529 0.132 0.339 9.624 38.27
Total 55.584150 1.000000 3 1.199 0.277 0.487 13.799 33.54
4 2.045 0.433 0.616 17.453 31.93
5 3.108 0.595 0.724 20.706 31.47
6 4.426 0.759 0.811 22.187 33.63
The output from phase 1 = Θ1 = η1, μ1 = 0.616 x 0.3333
= 0.205 trucks/min
Lq1 = average number of trucks waiting in the queue at
the shovel,
= 1 x ∑ (state 10, 16, 19, 26, 29, 33 probabilities)
+ 2 x ∑ (state 20, 30, 34 probabilities)
+ 3 x ∑ (state 35 probability)
= 0.42 trucks
Lq3 = average number of trucks waiting in the queue at
the dump,
= 1 x ∑ (state 5, 12, 13, 24, 25, 26 probabilities)
+ 2 x ∑ (state 11, 22, 23 probabilities)
+ 3 x ∑ (state 21 probability)
= 0.09 trucks
Wq1 = average waiting time in the queue at the loader,
= Lq1/Θ = 2.045 min
Wq3 = average waiting time in the queue at the dump,
= Lq31/Θ = 0.433 min Figure 4—Unit cost vs. number of trucks
▲

The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 109 JULY 2009 437
text:Template Journal 7/17/09 1:25 PM Page 438

Optimization of shovel-truck system for surface mining

Table VI

Optimum truck solution

Path Optimum truck fleet size Waiting time(min) Utilization of shovel Production tons/min Unit cost cents/ton
At shovel At waste

S11-W5 3 2.48 0.54 0.781 22.137 20.91 **

S11-W6 4 2.59 0.53 0.709 20.095 27.73
S12-W5 6 2.15 0.99 0.686 28.683 27.04
S12-W6 5 2.09 0.99 0.785 29.69 22.94 **
S21-W5 6 2.24 1.03 0.701 29.292 26.48 **
S21-W6 6 1.81 0.86 0.625 26.157 29.65
S22-W5 4 2.47 0.51 0.690 19.563 28.48 **
S22-W6 5 3.11 0.59 0.724 20.706 31.47

formulation determines the optimal path for trucks to follow. W5 (waste 5), from S12 to W6, from S21 to W5 and from
We assume that path to follow for a truck does not change in S22 to W5. This result is in close agreement with the queuing
time, as in real time dispatching. network solution. Figure 5 shows the optimal dispatching
If we call Xij the average number of trucks per minute paths. When one examines the optimal paths, they are the
over path i-j at an instant snapshot of the system, path same paths which queuing solution results with minimum
variables are as shown in Table VII. loading and hauling costs.
The following LP formulation determines the optimal The optimizing study for Orhaneli open pit mine results
routes for trucks. in producing about 10.1 million m3 overburden removal in a
The objective function minimizes the total number of year with 4 shovels (2 units of 15 yd3 and 2 units of 10 yd3)
trucks, the number of trucks on the road, the number of and 18 units of 77 tons trucks (which is the objective
trucks at the shovels, and the number of trucks at waste function value) over the required minimum 9 million m3
dumps. Such as: yearly overburden removal. This analysis does not include
Min Z = number of trucks on the road * travelling time equipment breakdown. The average cost of hauling is 19.07
over that path + ¢/m3
number of truck at sink points (incoming) *
duration at that point + number of trucks at
source points (shovels) Table VII
MIN = 2.5*X15+1.5*X51+5.4*X16+3.9*X61+6.5*X25+4.6
Path variables for LP modelling
*X52+4.7*X26+3.0*X62+6.0*X35+4.6*X53+8.0*X
36+5.3*X63+5.7*X45+4.2*X54+7.5*X46+5.0*X64 Path Path variable Path Path variable
+1.5*X15+1.5*X25+1.5*X35+1.5*X45+1.5*X16+1.
S11-W5 X15 W5-S11 X51
5*X26+1.5*X36+1.5*X46+4;
S11-W6 X16 W6-S11 X61
Subject to: S12-W5 X25 W5-S12 X52
Balancing equations at each node (incoming-outgoing = S12-W6 X26 W6-S12 X62
S21-W5 X35 W5-S21 X53
0); S21-W6 X36 W6-S21 X63
X51+X61–X15–X16 = 0; S22-W5 X45 W5-S22 X54
X52+X62–X25–X26 = 0; S22-W6 X46 W6-S22 X64

X53+X63–X35–X36 = 0;
X54+X64–X45–X46 = 0;
X15+X25+X35+X45–X51–X52–X53–X54 = 0; Table VIII
X16+X26+X36+X46–X61–X62–X63–X64 = 0;
Result of LP problem
Limiting rates at sources (truck rates being processed at
source points, i.e. ∑outgoing = 1/loading time): Variable Value
X15+X16 = 1/3;
X25+X26 = 1/2.033; X15 0.3333
X16 0.0000
X35+X36 = 1/2.033; X25 0.0000
X45+X46 = 1/3; X26 0.4918
X35 0.4918
Nonnegativity constraints; X36 0.0000
(X15,X51,X16,X61,X25,X52,X26,X62,X35,X53,X36,X6 X45 0.3333
3,X45,X54,X46,X64)>=0 X46 0.0000
X51 0.3333
The result of the above LP formulation (Table VIII) shows X61 0.0000
that the optimum path for trucks should be such that nonzero X52 0.0000
X62 0.4918
values and path capacities are in trucks/min for each valid
X53 0.4918
path. Figure 6 illustrates the optimum paths as determined by X63 0.0000
the LP model for a given set of travel times and shovel X54 0.3333
X64 0.0000
loading times. The optimal paths are: from S11 (shovel 11) to
▲

438 JULY 2009 VOLUME 109 The Journal of The Southern African Institute of Mining and Metallurgy
text:Template Journal 7/17/09 1:25 PM Page 439
Optimization of shovel-truck system for surface mining
Figure 5—Optimum truck routes for dispatching
Total production = 100.682 tons/min
= 50x100.682 = 5034 tons /h
(assuming 50 minutes work per hour)
= 12x5034 = 60,408 tons/day (12h per
working day)
= 300x60,408=18 122 400 tons/year
(300 working days per year)
= 14,048 372 m3/year (loose)
= 10 106 743 m3/year (in place)
Average cost = 24.60 ¢/ton or 19.07 ¢/m3
Conclusion
The methodologies developed and presented in this paper
have the potential to be useful for mine operators for loading
and haulage planning in open pit mines and/or at the stage
of equipment procurement. Since the cost of shovels and
trucks is several hundred dollars per hour, the application of
the methodologies has potential for substantial savings. The
methodologies developed have been validated for a range of
shovels and off-highway dump trucks. The process has
proven the applicability of the theoretical model proposed by
the authors.
The first stage consisted of determining of the optimal
number of trucks working with each shovel in the system
using a model based on the closed queuing network theory.
A complete example has been provided for shovels working
with identical trucks. The results clearly demonstrate the
applicability of such an approach for the issues under study.
As a result of the queuing network solution, the optimum
truck number, which minimizes the unit cost hauled for
possible paths along with shovel utilization and
production/minute ise found to be: from S11 (shovel 11) to
W5 (waste 5) with 3 trucks, from S12 to W6 with 5 trucks,
from S21 to W5 with 6 trucks and from S22 to W5 with 4
trucks, which result in lower costs using 18 trucks in total.
At the next stage, it has been determined how the trucks
should be dispatched to shovels, using the LP model. Results
obtained are interesting and applicable to planning loading
and haulage operations in open pit mines or at the
procurement stage of the equipment. The optimal route of
trucks for Orhaneli coal mine is: from S11 (shovel 11) to W5
The Journal of The Southern African Institute of Mining and Metallurgy
T
(waste 5), from S12 to W6, from S21 to W5 and from S22 to e
W5. This result is in close agreement with the queuing
network solution, which provided the minimum loading and
c
hauling costs. h
n
Acknowledgements
i
This work was supported by the Research Fund of Istanbul
University. Project number: 62/23012003 and UDP- c
887/05122006. a
l
Appendixes
List of symbols
N
N number of cycling units (trucks)
M number of service centers o
μI service rate at i th phase t
Lqi expected number of trucks e
Θ number of trucks being serviced at the i’th phase
during one unit time
Wi expected time that a truck spends in the i’th phase
C1 cost per unit time of shovel
C2 cost per unit time of a truck
C total cost for unit production
References
ALARIE, S. and GAMACHE, M. (2002), Overview of solution strategies used in
truck dispatching systems for open pit mines, International Journal of
Surface Mining, Reclamation and Environment, vol. 16, no. 1, 2002.
pp. 59–76.
BASCETIN, A. An Applıcatıon of the Analytic Hierarchy Process in Equipment
Selection at Orhaneli Open Pit Coal Mine, Transactions of the Institutions
of Mining and Metallurgy-(AusIMM), Section A-Mining Technology.
vol. 113, no. 3, 2004. pp.192–199(8).
BASCETIN, A. Investigation of Costs of an Open Pit Mine by the 3D-Modelling of
Mine Basin, Scientific Research, Istanbul University Research Fund, 2002,
Project No : 1607/30042001.2002.
BARNES, R.J., KING, M.S., and JOHNSON, T.B. Probability Techniques for
Analyzing Open Pit Production Systems, Application of Computers and
Operations Research in the Mineral Industry (16th APCOM Symposium),
1978. pp. 462–476.
CARMICHAEL, D.G. Shovel-truck queues: a reconciliation of theory and practice,
Constr. Mgmt. And Economics, vol. 4, 1986. pp. 161–177.
CARMICHAEL, D.G. Engineering Queues in Construction and Mining, Ellis
Horwood Ltd., Toronto, Canada, 1987.
ELBROND, J. Haulage system analysis. Queuing theory, Surface Mining, 1990.
pp. 743-748.
KOENIGSBERG, E. Twenty Five Years of Cyclic Queues and Closed Queue
Networks, A Review, Journal of Operational Research Society, vol. 33,
1982. pp. 605–619.
MUDULI, P.K. and YEGULALP, T.M. Modeling truck-shovel systems as closed
queuing network with multiple job classes, International Transactions in
Operational Research, vol. 3, no. 1, 1999. pp. 89–98.
NEONEN, L.K., GRAEFE, P.W.U., and CHAN, A.W. Interactive computer model for
truck/shovel operations in an open-pit mine, Winter Simulation
Conference. Proceedings of the 13th conference on Winter simulation,
Atlanta, Georgia, vol. 1, 1981. pp. 133–139.
RAMANI, R.V. Haulage system simulation analysis in surface mining, SME
publications, B.A. Kennedy, (ed.), 1990. p. 724–742.
SABAH, A., KHALIL, M., and MOHAMED, A.A. Computer model for selecting
equipment for earthmoving operations using queuing theory, Construction
Informatics Digital Library available at: http://itc.scix.net paper w78-
2003-1.content. 2003.
SGUREV, V., VASSILEV, V., DOKEV, N., GENOVA, K., DRANGAJOV, S., KORSEMOV, C., and
ATANASSOV, A. An automated system for real-time control of the industrial
truck haulage in open-pit mines, Institute of Industrial Cybernetics and
Robotics, Sofia, Bulgaria, 2003. p. 1113.
SHANGYAO, Y., WEISHEN, L., and MAONAN, C. Production scheduling and truck
dispatching of ready mixed concrete, Transportation Research: Part E
(1366-5545), vol. 44, no. 1, 2008. pp. 164-179.
SOUMIS, F., ETHIER, J., and ELBROND, J. Evaluation of the new truck dispatching in
the Mount Wright mine, Application of Computers and Operations
Research in the Mineral Industry (27th APCOM Symposium), 1989.
pp. 674–682.
WHITE, J.W., ARNOLD, M.J., and CLEVENGER, J.G. Automated Open Pit Truck
Dispatching at Tyrone, Engineering and Mining Journal, vol. 6, 1982.
pp. 76–84. ◆
▲
VOLUME 109 JULY 2009 439